Match and Unify Poor Quality Foreign Data at Scale with 360Science in-memory API. The 360Science AI Engine is simply the world’s fastest and most capable engine for unifying global customer data – with the ability to accurately link millions of customer records in different data systems in just seconds.
The AI Engine treats address lines as an object, and does not require addresses to be validated, standardized or corrected prior to matching. Because the Engine has scoring engine separate from the matching logic, it is tolerant to significant variations in the data and can score the granular differences in inputs. This makes 360Science Matching Engine ideal for applications involving fraud detection or international data where address data is expected to be poorly formatted and uncorrectable.
The 360Science Matching Engine delivers the ability to easily match worldwide data and supports *all* character sets and *all* worldwide languages. Unicode input data is processed by the core engine, and transliterated into English Latin characters and then by the AI for matching. So, whether your data is Latin (Worldwide/Localized), Cyrillic Россия, Hellenic Ελλάδα, Hebrew ישראל, KANJI 日本, Simplified Chinese 中国, Arabic العربية, Thai ประเทศไทย, Hangul 대한민국, and even Klingon ˈt͡ɬɪŋɑn (yes – no kidding), the 360Science Matching Engine has you covered. When when paired with our global address validation engine address data capabilities are further enhanced.
Note: Transliteration functions are not supported in the mDesktop workstation application